1. Field of the Invention
The present invention relates to a method for service discovery in a mobile ad-hoc network and a system thereof. More particularly, the present invention relates to a method for a hybrid service discovery adapting between push-based and pull-based schemes in a mobile ad-hoc network and a system thereof, which makes it possible to use a service in the mobile ad-hoc network in a manner that a node that provides the service transmits a control packet by a push-based method and a node that intends to use the service transmits the control packet by a pull-based method.
2. Description of the Related Art
A mobile ad-hoc network (hereinafter referred to as ‘MANET’) is a network autonomously constructed among nodes having mobility without the help of the existing environments. All nodes in MANET are mobile, and thus the network is characterized by a very dynamic and flexible structure. When using such a MANET, communications among users can be supported without the existing Internet-based resources and the distance limitations for communications can be eliminated by a multi-hop support. According to these characteristics, MANET has an advantage that it can support users' communications any time and anywhere. Accordingly, MANET has attracted a great deal of interest for its potential as the core network technology to cope with ubiquitous environments in the future.
Service discovery schemes for discovering and automatically setting diverse network resources, for example, various kinds of devices such as network printers, video appliances, and television receivers, and various kinds of servers such as FTP servers and streaming servers, dispersed on a network, have become a matter of growing concern. These schemes have great advantages in that even a user who has no expert knowledge of networks can conveniently use network resources. Particularly, since it is very difficult for a user to manually perform a setting operation in a very dynamic ad-hoc network such as a MANET, it is much more necessary to provide the service discovery function in a MANET.
Accordingly, research for service discovery methods in MANET has been actively performed heretofore. However, since routing support in a very dynamic MANET environment has performance limits, the improvement of the service discovery method is quite restricted.
The service discovery method on MANET can be divided into a centralized method and a decentralized (that is, distributed) method in accordance with existence or nonexistence of a centralized service manager.
In the centralized service discovery method, a separate server that is a service directory for providing kinds of devices and their IDs available in the network is provided. A node that can provide services such as printing and so on (hereinafter referred to as a ‘service provider’) registers kinds of services and their IDs provided by the node itself in the service directory. A node that intends to use the services (hereinafter referred to as a ‘service demander’) accesses the corresponding server and discovers a service provider that can provide the service desired by the service demander. However, since this method has a network structure seriously depending on the service directory, it is not suitable to the MANET environment. Particularly, if the service directory server itself moves from MANET and becomes unavailable, all nodes in the network cannot use the services.
In contrast, in the decentralized service discovery method, a separate directory server is not provided but every node in the network obtains information about a service provider. The decentralized service discovery method is divided into a push-based method and a pull-based method in accordance with the subject of transaction.
According to the push-based method, a service provider periodically transmits its information. A service demander stores such information, and if the corresponding service provider is needed, the service demander discovers and uses the service provided by the service provider. In this method, since the service information should be periodically transmitted to the entire network, a large number of control packets are produced in proportion to the number of service providers. This causes the performance of the entire network to deteriorate.
According to the pull-based method, a service demander itself discovers a service provider by transmitting a request message to the entire network at a time point when the corresponding service is needed. This method can reduce the number of control packets in the network in comparison to the push-based method. However, a process of discovering the service provider is additionally needed, and this causes a time delay in using the service.